Research on Association Rules Mining Base on Positive and Negative Items of FP-tree

摘要

This paper analyzes the two kinds of classic traditional association rules algorithm Apriori and FP - tree which those are advantages and disadvantages. In order to solve the Apriori algorithm for mining association rules requiring multiple scanning database and generate a large number of candidate frequent sets, base on the inherited FP -tree algorithm the advantages of scanning second times of database project on the basis of combining the positive and negative association rules mining algorithm, not only can solve real problems in negative association rules mining, can also delete the contradictory relationship, effectively improve the efficiency of the algorithm. The final process of a set of real transaction of mining experiments show that the algorithm to improve the quality and efficiency of mining rules, and to avoid invalid mode rules.